Progress toward the clinical application of gene therapy for arthritis has been slowed by the inability to sustain intraarticular transgene expression. The prevailing observation is that transgene expression persists for only two to three weeks. The current proposal is designed to test the following hypotheses: 1) The immunologic incompatibility of foreign (non-self) transgene products leads to this premature termination of transgene expression, 2) Specific cell types within the synovium are capable of supporting stable, persistent transgene expression, and 3) In normal, immunocompetent animals, expression of cDNAs encoding homologous (self) or immunologically compatible gene products will persist long-term within the synovium. To determine the relative benefits and limitations of various ex vivo and in vivo gene delivery systems, each will be evaluated for transgenic persistence in the knees of athymic, nude rats. Because transduced cells that express foreign proteins survive for their natural life-span in this animal, it will serve as a "pseudohomologous" system within which we can use diagnostic marker genes to simulate the delivery of transgenes encoding "self" proteins to normal joints. Using the cDNA for Green Fluorescent Protein to phenotypically tag genetically modified cells in the synovium, we will use flow cytometry to characterize the populations of transduced cells. By studying how they change with time, we will identify the specific cell populations within the synovial lining that may permit long-term expression. The true motivation behind the study is however to determine for how long exogenous transgenes may be expressed in normal and arthritic joints in immunocompetent animals. For this, two different gene systems will be studied in normal rats that will not activate immune clearance. The first will be a soluble form of the rat TNF receptor type II; the second will be a "gene" construct that is transcribed, but not translated, avoiding altogether the immune issues of foreign protein products. The following specific aims will be addressed: 1) To comparatively evaluate synovial fibroblasts, dermal fibroblasts and mesenchymal stem cells for their capacity to enable persistent transgenic expression, and their biodistribution when used as vehicles for ex vivo intra-articular gene delivery to the knees of athymic nude rats, 2) To comparatively evaluate recombinant lentivirus, AAV and high capacity adenovirus for their capacity to enable persistent transgenic expression and their biodistribution when used as vehicles for direct intraarticular gene delivery to the knees of athymic nude rats, 3) To characterize genetically modified cell populations in the synovium and to determine how they change with time in vivo, and 4) Within immunocompetent rats, evaluate persistence of expression using immunocompatible transgene systems in normal and arthritic knees following ex vivo and in vivo delivery.